Implanted pacemakers straighten out a wobbly heart rhythm. Implanted cardioverter-defibrillators (ICDs) jump-start a helplessly fluttering heart. These state-of-the-art electronic devices save lives. But they're complex machines that can -- and do -- break down.
What's the risk of this happening? And when manufacturers announce recalls of problem ICD models, what's the risk of having your implant replaced?
The new studies now open a window on this crucial information. They appear in the April 26 issue of the Journal of the American Medical Association, along with an editorial by Bruce L. Wilkoff, MD, director of cardiac pacing and tachyarrhythmia devices for The Cleveland Clinic.
"These devices have improved in reliability, but people are just now becoming aware of the fact they are not perfect," Wilkoff tells WebMD. "This is a small problem that solves a very big problem. Cardiac deaths are the most common reason people die in the U.S. today. We have an incredibly successful treatment, but there are costs both in money and in morbidity."
The Risk of Pacemaker, ICD Failure
William H. Maisel, MD, MPH, director of the pacemaker and device service at Boston's Beth Israel Deaconess Hospital in Boston, analyzed device malfunctions reported to the U.S. FDA. He also analyzed pacemaker and ICD implant patients followed in three different studies in the U.S., the U.K., and Denmark.
"We wondered whether the increasing complexity of these devices has been associated with decreased reliability," Maisel tells WebMD.
Maisel and colleagues looked at reports of pacemaker and ICD malfunctions reported to the FDA between 1990 and 2002. During this time, there were 2.2 million pacemakers in use, and more than 415,000 ICDs. There were 17,000 reports of devices that had to be replaced. Hardware problems caused most of the breakdowns.
Reassuringly, the annual rate of pacemaker replacement due to malfunction went down, suggesting these devices became more reliable," Maisel says. "ICD malfunction declined during the first part of study, from 1994-1998, then increased from 1998-2002. This suggests that ICDs became less reliable as they became more complicated."
The usual problem with FDA data is that the FDA adverse-event reporting system is voluntary. Might missing reports have skewed the data?
"Combined, the studies included patients implanted with several hundred thousand pacemakers and thousands of ICDs," Maisel says. "The results were amazingly consistent with the FDA reports. They also showed that the pacemaker malfunction rate has declined significantly. The ICD malfunction rate also declined notably from the 1980s to the mid-1990s, but again we saw an increase in the ICD malfunction rate from 1990 to 1992. The rate of malfunction for ICDs increased more than fourfold from 1998-2001."
There is good news, however. New data from 2003 and 2004 show what Maisel calls "a reassuring decline in the ICD malfunction rate, almost back down to the lowest levels seen in the 1990s."
Does the risk of a breakdown mean you should think twice about getting a pacemaker or ICD device? Maisel doesn't think so -- if your doctor says you really need one.
"The chance someone will be saved by one of these device about a thousand times greater than the chance the device will fail when they need it," Maisel says.
He notes that there were 30 deaths due to pacemaker malfunction and 31 deaths due to ICD malfunction among the 2.7 million patients who had one of these devices.
ICD Replacement Risks
ICD breakdown isn't the only risk of getting one of these implants, find cardiologist Andrew D. Krahn, MD, of University of Western Ontario in London, Canada, and colleagues.
When a device breaks down, manufacturers sometimes discover a design flaw or manufacturing error. That leads to product "advisories" that warn patients of potential problems. This puts the patient in a major dilemma. Should patients risk device failure, or should they risk the operation to have the device removed and replaced?
"We saw the other side of the coin: not devices failing, but patients suffering due to recall of devices," Krahn tells WebMD.
Getting an ICD implant isn't a major operation. But replacement can be tricky. The patient is older, there may be scar tissue, and there is more risk of a deadly infection after surgery. How dangerous is it? Krahn's team surveyed nearly all the doctors in Canada who do these procedures.
"We were surprised by the high risk of infection and the number of replacement procedures that required a second operation," Krahn says. "We reported a 6% major complication rate. And 2% of patients with major complications required a major surgical procedure. That is a compelling risk. Our concern about the device has to cap that threshold for us to advocate replacing it."
So what should you do if you've received an "advisory" warning that your ICD might fail? This is a tough question, and the answer may be not to have the device removed.
"The majority of advisories had a 1-in-10,000 to a 1-in-100 risk of device failure," Krahn says. "And for some manufacturer advisories, you can detect problems by monitoring the device. So the vast majority of the time, you detect a problem before the patient feels anything. On the other hand, some advisories warn of abrupt loss of function. Some patients suffer sudden death."
It's an unusual risk-vs.-benefit decision. Patients, Krahn says, should first look at their own medical condition, then at the risk estimated for the specific device they have implanted.
"Let's say a man had a first heart attackheart attack three years ago, and the doctor advised him that his risk of sudden death is 2% to 5% per year," Krahn says. "If that patient has device put in and finds out there is a 1-in-1,000 chance the device will not be reliable, the patient may say leave that alone. If, on the other hand, that patient received a shock three months ago, and the doctor says, 'We checked the device, it saved your life,' then the patient may want a new one, because his life clearly depends on it working."
Wilkoff agrees that patients must carefully consider whether to replace an ICD because of an advisory.
"It turns out that changing the device is not necessarily the lower-risk answer," he says. "We shouldn't overreact to the realization these remarkably effective devices aren't perfect. Maisel found a 1%-2% malfunction rate, not death risk. But in the ICD population, there is a 7% to 15% annual risk of death without a device. Is the ICD perfect? No. But we cannot have people panicking about this when there are much bigger devils sitting outside their doors."